(navigation image)
Home American Libraries | Canadian Libraries | Universal Library | Community Texts | Project Gutenberg | Children's Library | Biodiversity Heritage Library | Additional Collections
Search: Advanced Search
Anonymous User (login or join us)
Upload
See other formats

Full text of "Enlistment standards as related to performance in Aviation Antisubmarine Warfare Operator and Aviation Antisubmarine Warfare Technician ratings"

NAVAL POSTGRADUATE SCHOOL 

Monterey, California 




THESIS 




ENLISTMENT STANDARDS AS RELATED TO 
PERFORMANCE IN AVIATION ANTISUBMARINE 
WARFARE OPERATOR AND AVIATION 
ANTISUBMARINE WARFARE TECHNICIAN RATINGS 


by 




Clyde D. Sandel 




and 




Mary F. Gleason 




September 1983 




Thesis Co-Advisors: R. S. 

W. E. 


Elster 
McGarvey 



Approved for public release, distribution unlimited 



Unci ^ssi fjUSd 



SECURITY CLASSIFICATION OF THIS PAGE (Whan Data Entered) 



REPORT DOCUMENTATION PAGE 



READ INSTRUCTIONS 
BEFORE COMPLETING FORM 



1. REPORT NUMBER 



2. GOVT ACCESSION NO 



3. RECIPIENT'S CATALOG NUMBER 



4. TITLE (and Suotltle) 

Enlistment Standards as Related to 
Performance in Aviation Antisubmarine 
Warfare Operator and Aviation Anti- 
submarine Warfare Technician Ratings 



5. TYPE OF REPORT & PERIOD COVERED 

Master's Thesis 
September 1983 



6. PERFORMING ORG. REPORT NUMBER 



7. AuTHORf*; 

Clyde D. Sandel and Mary F. Gleason 



8. CONTRACT OR GRANT NUMBERf*} 



t. PERFORMING ORGANIZATION NAME ANO AOORESS 

Naval Postgraduate School 
Monterey, California 93943 



10. PROGRAM ELEMENT PROJECT, TASK 
AREA 4 WORK UNIT NUMBERS 



11. CONTROLLING OFFICE NAME ANO AOORESS 

Naval Postgraduate School 
Monterey, California 93943 



12. REPORT DATE 

September 1983 



13. NUMBER OF PAGES 

74 



14. MONITORING AGENCY NAME ft AODRESSfif dlttarant from Controlling Office) 



15. SECURITY CLASS, (ot thie report) 

Unclassified 



15a. DECLASSIFICATION/ DOWNGRADING 
SCHEDULE 



!«. DISTRIBUTION STATEMENT (ol thle Raport) 

Approved for public release, distribution unlimited. 



17. DISTRIBUTION STATEMENT (at tne abattmct entered In Block 30, It dlttarant from Raport) 



IS. SUPPLEMENTARY NOTES 



IS. KEY WOROS (Continue on ravataa tide It naeaaaary arm identity by block number) 

AX and AW Ratings Enlistment Standards, Enlistment Standards 
for AX and AW Ratings. 



20. ABSTRACT (Cantlnua on rararaa ttde it naeaaaary and Identity by block number) 

The purpose of this study is to discover if the Navy's system 
of assigning personnel to the Aviation Antisubmarine Warfare 
Technician (AX) and the Aviation Antisubmarine Warfare Operator 
(AW) ratings can be improved. A multivariable model is developed 
using "success" and "failure" as criterion variables. Bio- 
graphical and aptitude data available at the time of enlistment 
are used as predictor variables. Two indeoendent models were 



D0,^ 



FORM 

AN 71 



1473 EDITION OF 1 NOV SS IS OBSOLETE 

S/N 0102- LF- 014- 6601 



Unclassified 

SECURITY CLASSIFICATION OF THIS PAGE (When Date Bntarec 



Lla&j ^ qi " f i *^ 



SECURITY CLASSIFICATION OF THIS PAGE (Whan D«« £n«.r.d> 



Block 20 (continued) 

created using data available on personnel entering the Navy in 
1976, 1977 and 1978. The models were then validated on a new 
sample. 

These models predict the future fleet performance of AX and 
AW personnel as measured by length of service, paygrade achieved, 
and recommendation for reenlistment . Other results and recom- 
mendations regarding implementation and future research are 
discussed. 



S'N 0102- LF-014-6601 2 

Unclassified 



SECURITY CLASSIFICATION OF THIS RAGEfWh»n Dmtm Enfrmd) 



Approved for public release, distribution unlimited 



Enlistment Standards as Related to Performance in 

Aviation Antisubmarine Warfare Operator and 
Aviation Antisubmarine Warfare Technician Ratings 



by 



Clyde D. Sandel 
Lieutenant Commander, United States Navy 
B.A., Angelo State University, 1973 

and 

Mary F. Gleason 
Lieutenant, United States Navy 
B.S., University of Southern Mississippi, 1973 



Submitted in partial fulfillment of the 
requirements for the degree of 

MASTER OF SCIENCE IN MANAGEMENT 

from the 

NAVAL POSTGRADUATE SCHOOL 
September 1983 



■ 



ABSTRACT 



The purpose of this study is to discover if the Navy's 
system of assigning personnel to the Aviation Antisubmarine 
Warfare Technician (AX) and the Aviation Antisubmarine Warfare 
Operator (AW) ratings can be improved. A multivariate model 
is developed using "success" and "failure" as criterion 
variables. Biographical and aptitude data available at the 
time of enlistment are used as predictor variables. Two 
independent models were created using data available on 
personnel entering the Navy in 1976, 1977 and 1978. The models 
were then validated on a new sample. 

These models predict the future fleet performance of AX and 
AW personnel as measured by length of service, paygrade achieved, 
and recommendation for reenlistment . Other results and recom- 
mendations regarding implementation and future research are 
discussed. 



TABLE OF CONTENTS 

I. INTRODUCTION 3 

II. DATA BASE DEVELOPMENT 11 

III. THE VARIABLES 13 

A. BACKGROUND 13 

B. CRITERION VARIABLES 17 

C. PREDICTOR VARIABLES 19 

IV. STATISTICAL TECHNIQUES 21 

A. FREQUENCY ANALYSIS 21 

B. MULTIVARIATE CORRELATION ANALYSIS 21 

C. STEPWISE REGRESSION 22 

D. DISCRIMINANT ANALYSIS 22 

V. MODELS 24 

A. AX MODEL 24 

B. AW MODEL 26 

C. ADDENDUM 27 

VI. CONCLUSIONS 29 

LIST OF REFERENCES 70 

BIBLIOGRAPHY 72 

INITIAL DISTRIBUTION LIST 74 



LIST OF TABLES 

1. INTER-SERVICE SEPARATION CODE FOR THE AW RATING 31 

2. INTER-SERVICE SEPARATION CODE FOR AX RATING 32 

3. PREDICTOR VARIABLES 33 

4. AW ASVAB APTITUDE AREA SCORE — SUBSCALE GI 34 

5. AW ASVAB APTITUDE AREA SCORE — SUBSCALE NO 35 

6. AW ASVAB APTITUDE AREA SCORE — SUBSCALE AD 36 

7. AW ASVAB APTITUDE AREA SCORE — SUBSCALE WK 37 

8. AW ASVAB APTITUDE AREA SCORE — SUBSCALE AR 33 

9. AW ASVAB APTITUDE AREA SCORE — SUBSCALE SP 3 9 

10. AW ASVAB APTITUDE AREA SCORE — SUBSCALE MK 40 

11. AW ASVAB APTITUDE AREA SCORE — SUBSCALE EI 41 

12. AW ASVAB APTITUDE AREA SCORE — SUBSCALE MC 42 

13. AW ASVAB APTITUDE AREA SCORE — SUBSCALE GS 4 3 

14. AW ASVAB APTITUDE AREA SCORE — SUBSCALE SI 44 

15. AW ASVAB APTITUDE AREA SCORE — SUBSCALE AI 4 5 

16. AW SCREEN SCORE 46 

17. AW RACE DISTRIBUTION 47 

18. AW ENTRY PAY GRADE (E00-011) 47 

19. AW MARITAL STATUS/DEPENDENTS 47 

20. AW AFQT SCORE FREQUENCY 48 

21. AW TERM OF ENLISTMENT (NO. OF YEARS) 4 9 

22. AX ASVAB APTITUDE AREA SCORE — SUBSCALE GI 4 9 

23. AX ASVAB APTITUDE AREA SCORE — SUBSCALE NO 50 



24. AX ASVAB APTITUDE AREA SCORE — SUBSCALE AD 51 

25. AX ASVAB APTITUDE AREA SCORE — SUBSCALE WK 52 

26. AX ASVAB APTITUDE AREA SCORE — SUBSCALE AR 53 

27. AX ASVAB APTITUDE AREA SCORE — SUBSCALE SP 54 

28. AX ASVAB APTITUDE AREA SCORE — SUBSCALE MK 55 

29. AX ASVAB APTITUDE AREA SCORE — SUBSCALE EI 56 

30. AX ASVAB APTITUDE AREA SCORE — SUBSCALE MC 5 n 

31. AX ASVAB APTITUDE AREA SCORE — SUBSCALE GS 58 

32. AX ASVAB APTITUDE AREA SCORE — SUBSCALE SI 5 9 

33. AX ASVAB APTITUDE AREA SCORE — SUBSCALE AI 6 

34. AS SCREEN SCORE 61 

35. AX RACE DISTRIBUTION 62 

36. AX ENTRY PAY GRADE (E00-011) 62 

37. AX MARITAL STATUS/DEPENDENTS 62 

38. AX AFQT PERCENTILE (OR EQUIVALENT) 6 3 

39. AX TERM OF ENLISTMENT (NO. OF YEARS) 64 

40. AX STEPWISE SELECTION: SUMMARY 

TERMENLT AS A VARIABLE 64 

41. AX STEPWISE SELECTION: SUMMARY 

WITHOUT TERMENLT 64 

42. AX DISCRIMINANT ANALYSIS 65 

43. AW STEPWISE SELECTION: SUMMARY 

WITH TERMENLT 66 

44. AW STEPWISE SELECTION: SUMMARY 

WITHOUT TERMENLT 66 

45. AW DISCRIMINANT ANALYSIS 67 

46. AX DISCRIMINANT ANALYSIS 68 

47. AW DISCRIMINANT ANALYSIS 69 



I. INTRODUCTION 

The objective of this study is to discover if selection 
standards for Aviation Antisubmarine Warfare Technicians (AX) 
and Aviation Antisubmarine Warfare Operators (AW) can be 
improved by utilizing data available at the time of enlistment. 
Studies concerning personnel assignments to ratings have 
traditionally used training criteria, with completion of Class 
A School as the measure of success for validation [Ref. 1]. 
Other studies have focused on whether or not an individual 
leaves the service as the measure of success. This study will 
use measures of the operational performance of AX's and AW's 
in the fleet as the dependent variables. 

The following discussion provides a brief overview of each 
rating . 

AX - The AX rating is responsible for keeping aviation 
antisubmarine warfare (ASW) weapon systems and system components 
operating in good condition. As such, the training for the 
rating is of a highly technical nature. The AX community is 
relatively small and is unique to those Naval squadrons whose 
principal purpose is air antisubmarine warfare. Such squadrons 
consist of the S-3, P-3, HS and HSL. these squadrons' operational 
mission effectiveness is directly linked to the performance and 
quality of the members of the AX rating. AX's perform in-flight 
maintenance of airborne electronic systems, remove and install 



units of ASW equipment, maintain operating efficiency of ASW 
equipment, perform a wide range of electronic shop operations, 
debrief flight crews, and read and apply equipment service 
diagrams, schematics and manuals. Important qualifications 
for the AX rating include manual dexterity, arithmetic ability 
and an ability to do detail work [Ref. 2]. 

AW - The AW rating is comprised of two components, AWA 
(Acoustics Operators) and AWH (Non-Acoustic Operators). For 
the purpose of this study, the term AW will include both 
components. AW's operate airborne radar and electronic 
equipment used in detecting, locating and tracking submarines. 
They also operate radar to provide information for aircraft 
and surface ship navigation. Some individuals may also act as 
helicopter rescue crewmen. They work as part of the flight 
crew on long range and intermediate range aircraft and on 
helicopters. Again as with the AX rating, AW's play a key 
part in a squadron's operation mission effectiveness. Important 
qualifications for the AW rating include manual dexterity and 
competence with tools, equipment and machines, good arithmetic 
and record-keeping ability and the ability to do intricate 
work and repetitive tasks [Ref. 3]. 

With the advent of the All Volunteer Force, a projected 
growth to a 600 ship Navy, increasing costs, both in equipment 
and in personnel, and a decline in the 17-21 year old male 
population, the need to study and refine enlistment standards 
and assignment techniques is obvious [Ref. 4]. 



A study by Thomason [Ref. 5] indicated that first term 
attrition among Navy recruits is dependent upon initial rating 
assignments. This finding, combined with the aforementioned 
reasons, prove the need for further studies and research in 
the area of assignment techniques. Better assignment techniques 
and selection processes should result in lower training costs, 
improved readiness, higher retention and a more experienced, 
effective Navy. 



10 



II. DATA BASE DEVELOPMENT 

Information on over 206,000 personnel was compiled by 
merging: (1) the Defense Manpower Data Center ( DMDC ) Cohort 
File; (2) a Navy Health Research Center (NHRC) file; (3) a 
promotional advancement exam file; and (4) a Chief of Naval 
Education and Training (CNET) file. The DMDC Cohort File 
contains demographic variables obtained at the time of accession 
Additionally, it is updated quarterly with active duty informa- 
tion including information on separation from service if 
appropriate. Continuously updated, the NHRC file contains 
medical statistics on personnel from the date of enlistment to 
date of discharge. The CNET file includes advancement and 
training information. From this data base, information on 
1094 and 559 non-prior service personnel associated with the 
AW and AX ratings, respectively, was extracted. 

By using the Statistical Analysis System ( SAS ) , a number 
of logic screens were implemented to eliminate data on 
individuals felt to be inappropriate for analysis because 
their separation did not reflect failure in the fleet opera- 
tional environment. Frequency distributions of inter-service 
separation codes (Tables 1 and 2) provide breakdowns explain- 
ing how personnel exited the Navy. Personnel with the follow- 
ing inter-service separation codes were specifically deleted: 



11 



Code Reason for Separation 

10 Medical conditions existing prior to service 

11 Medical disability with severence pay 

12 Permanent medical disability - retired 

13 Temporary medical disability - retired 

14 Medical disability without severence pay 

15 Medical disability - Title 10 retirement 

16 Unqualified for active duty - other 
22 Dependency or hardship discharge 

32 Death 

40 Entry into officer commissioning program 

41 Entry into warrant officer program 

42 Entry into service academy 
50 20-30 years of service 

94 Pregnancy 

As a result of applying the screens, 1048 and 405 AW s and 
AX's were identified as personnel appropriate for analysis. 
These groups were placed in separate data sets. One data set 
includes all personnel who began in the AW rating. Because 
some AX's were originally classified into an Avionics Technician 
(AV) rating, the other data set includes those personnel who 
initially began as AV's and were later classified as AX's as 
well as those personnel who began as AX's 



12 



III. THE VARIABLES 

A. BACKGROUND 

Current enlistment standards are based jointly on predicted 
recruit survival rates and on mental aptitudes. In actuality, 
survival rates have not always been an issue, and not until 
the early 1970' s did mental aptitude start receiving concentrated 
study [Ref. 6]. Clearly the reason that survivability is being 
extensively studied for its role in the selection and assignment 
process of Navy recruits is that by extending a recruit's 
survivability (reducing attrition), the Navy reduces training 
and replacement costs, and increases individual and unit 
performance. Mental aptitude is viewed as a key factor not 
only in survivability, but also in its role in the individual/ 
skill matching process. 

Studies dealing with survivability have analyzed survival 
rates at recruit training, Class A School, first term of 
enlistment, and from first through eight years of service 
[Ref. 7]. 

Predictor variables used are generally a composite of two 
or more of the following: (a) the Armed Forces Qualification 
Test (AFQT) which for ASVAB forms 5, 6, and 7 was a composite 
score based on three ASVAB subtests - Word Knowledge, Arithmetic 
Reasoning and Spatial Perception; (b) age; (c) years of education; 
(d) high school graduation versus non-high school graduation; 



13 



(e) high school diploma versus General Equivalency Diploma; 

(f) marital status; (g) number of primary dependents; (h) race; 
(i) sex; (j) residence at time of service entry; (k) location 
of recruit training; (1) rating assigned; and (m) Delayed Entry 
Program (DEP) enlistment. 

The following is a summary of a few of the studies on 
enlistment standards and assignment processes. 

Lurie [Ref. 8] used AFQT score, number of dependents, and 
years of education to predict the performance of the Ship's 
Serviceman (SH) and Electronics Technician (ETN) ratings. He 
found that for the SH rating, non-high school graduates with 
lower AFQT scores were promoted faster than those with higher 
scores, however AFQT score had no impact on survival. The AFQT 
score did not aid in predicting advancement or survival for 
members of the ETN rating. 

Lockman [Ref. 9], in a study to determine the different 
survival rates of Class A School graduates vice non-Class A 
School attendees (GENDETS) found that the Class A School 
graduates with 12 or more years of education had higher survival 
rates than those in the GENDET category with 12 or 
more years of education, but non-school eligible (<50 AFQT 
score), had the higher survival rate. Additional findings 
indicated that the majority of Class A schoolers: (a) had 12 
or more years of education; (b) were school eligible; (c) joined 
the Navy under the Delayed Entry Program (DEP); (d) and survived 
four years of service. The opposite held true for the GENDETS. 



14 



Lurie [Ref. 10], in a study of eight year survival rates, 
found that the most important variable related to survival was 
educational level. In terms of survival for Class A School 
attendees, the optimal age was 17 - 21 years old. An interest- 
ing finding was that for Class A School attendees, members in 
mental group I (>90 AFQT ) had the worst survival rate. For 
non-Class A School attendees there was a general upward trend 
in survival as mental test scores decreased. 

In another study by Lockman [Ref. 11] on the effects of 
joining the Delayed Entry Program ( DEP ) , it was determined that 
after controlling for recruit quality (as measured by the SCREEN 
score) and training guarantees, those who were in DEP for three 
or more months had the highest survival rates. 

Thomason [Ref. 12] found in his study on first term enlist- 
ment survival rates on 37 different Navy ratings that age, 
education, DEP enlistment, recruit training location, race, 
number of dependents, mental group and follow on tour assign- 
ments had varying degrees of significance in determining 
survivability. 

Marcus and Lockman [Ref. 13], in their work on analyzing 
alternative enlistment standards to increase the supply of 
Navy recruits by improving survivor prediction rates, used a 
somewhat different approach in their selection of predictor 
variables. Rather than using the Armed Forces Qualification 
Test (AFQT), they chose instead to use those ASVAB subtests 
not included in the computation of the AFQT score, i.e., MK , 



15 



MC, EI, AI and SI. The intent was to use different ASVAB 
subtests in lieu of AFQT when computing a recruit's SCREEN 
score. The second variable selected was whether or not a 
recruit required an enlistment waiver and the gravity of the 
waiver required. The third variable, educational quality, is 
rather complex in nature, and involved capturing or measuring 
variations in the quality of high school diplomas and equival- 
ency ( GED ) tests by geographic region. Finally, the fourth 
variable selected was Class A School attendance or apprentice- 
ship training. 

Their results indicated that no large improvement in 
survivability prediction would occur from using different 
ASVAB subtest scores in the SCREEN table. Small increases in 
supply would occur from expanding somewhat on certain enlist- 
ment waivers. Again, increases in supply would occur by 
adjusting eligibility requirements to allow for measures of 
GED quality. Lastly, they concluded that separate screening 
of Class A School and apprenticeship trainees had potential 
for cost savings to the Navy. The above mentioned increases 
in supply, of course, relate to the increased numbers recruited 
by changing the different policies regarding waivers and GEDs . 

Lockman and Lurie [Ref. 14], in their work on updating the 
Navy's Success Chances of Recruits Entering the Navy (SCREEN) 
table, used a different measure of education and mental aptitude 
The SCREEN table in use during their study was based on a 
composite score of grade of education, whether or not an 



16 



applicant had dependents, AFQT score and age. A minimum 
score of 70 was required for enlistment and the survival 
predictions were for the first year of service. They replaced 
highest grade of education with whether an applicant had a 
high school diploma (or more), certificate of equivalency 
(GED), or less than high school diploma. AFQT mental group 
(I, II, III, IV) replaced AFQT score. Results of their study 
indicated that by replacing the variables the SCREEN table 
could serve as a predictor of the entire first term of enlist- 
ment vice just the first year. 

Sands [Ref. 15], in a study to develop an instrument to be 
used by the Navy recruiters in the field to estimate an appli- 
cant's probability of surviving the initial two years of 
service, used ASVAB aptitude test scores, number of years of 
education, age and number of dependents as predictor variables 
His conclusion was that the model could be used effectively by 
recruiters and would produce reasonably accurate results. 

The above studies, although by no means all inclusive, 
indicate the key variables used in past research efforts. 

B. CRITERION VARIABLES 

This study defines "success" as: 

1. completion of 3.9 years of the initial term of 
enlistment, and 

2. achievement of paygrade E-4, and 

3. recommendation for reenlistment 



17 



"Failure" is achieved in this study if any, or a combination 
of any, of the following conditions were met: 

1. Failure to complete enlistment 

2. Failure to be recommended for reenlistment 

3. Failure to achieve paygrade E-4 

Category 1 in all tables and matrices denotes the "success" 
category. Category 2 in the various tables and matrices 
denotes the "failure" category. 

These two categories, "success" and "failure", are mutually 
exclusive but do not account for all of the AW s and AX's in 
the data set. Twenty-four personnel were excluded from AW 
analysis and sixteen were excluded from the AX analysis since 
they fell into a "gray area" in between the two criterion 
categories . 

The measures used in the success category are felt to be 
valid measures of success for first term enlistment. Even 
though recruits are enlisted on four or six year contracts, 
completion of three years and nine months was chosen as a 
measure of success because the cohort data were updated most 
recently in October 1982. The three years nine month measure 
is the longest period some of the 1978 recruits could have 
achieved. If the four or six year cutoffs had been used as a 
measure of success, many of those people who enlisted in the 
last three months of 1978 would have been incorrectly classified 
as failures. 



18 



Actual group membership of the 1976-1978 cohort groups is 
denoted below: 

Success Failure 

AX 235 154 

AW 665 308 

C. PREDICTOR VARIABLES 

Predictor variables were selected based on the past 
research discussed in the Background section of this thesis. 

The variables selected were measures of personal attributes 
that were know at the time of enlistment. 

The Navy currently uses SCREEN, AFQT, high school graduation, 
marital status and age as variables in the enlistment pocess. 
Additionally, Class A School eligibility (AFQT >49) and various 
ASVAB subtest scores are used in skill rating assignment. The 
ASVAB subtest scores used for the AX and AW ratings are as 
follows [Ref. 16]: 

AX AW 

MK+EI+GS =156 AR+2MK+GS = 200 
+AR = 218 

It should be noted that these formulae involve normed 
scores, while efforts in this study involve "raw", non-normed 
scores . 

By including Navy's current predictor variables in the 
analysis, a potential side benefit would be that of analyzing 
their effectiveness. 



19 



Eighteen predictor variables were selected for analysis in 
this study. Table 3 briefly identifies each variable and 
provides the number of the table containing the variable's 
frequency distribution. 



20 



IV. STATISTICAL TECHNIQUES 

The following is a brief description of the statistical 
procedures used in this analysis. 

A. FREQUENCY ANALYSIS 

Frequency distributions give a count of how frequently 
each value of the variables occurs among the data sets. In 
this study, frequency analysis was performed to provide the 
counts of "success" and "failure" as well as the counts of 
each predictor variable used in the models. Results are 
contained in Tables 4 through 21 for the AW s and Tables 22 
through 39 for the AX ' s . 

B. MULTIVARIATE CORRELATION ANALYSIS 

Through the use of this procedure the relationships between 
and among the variables have been studied. Casual interpreta- 
tion can not be made safely, but as a descriptive tool correla- 
tion analysis has potential for predicting values on one 
variable given information on another variable or set of 
variables. A summary measure that communicates the extent of 
relationship or correlation between a set of predictor variables 
and a criterion variable is called a multiple correlation 
coefficient, denoted by R. The value of the square of the R 
signifies the proportion of variance in the criterion variable 
predicted from the combined set of predictor variables. 



21 



C. STEPWISE REGRESSION 

Given a set of predictor variables, it is not necessary to 
utilize every one in the determination of a multiple R. Rather 
the stepwise regression procedure chosen begins by selecting 
the one predictor variable that correlates most highly with the 
criterion variable, and then introduces a second predictor 
variable, the one that accounts for the most of the remaining 
or residual variance in the criterion variable. Variables are 
continually added until inclusion of another predictor variable 
would account for only an insignificant amount of variance in 
the criterion variable. 

D. DISCRIMINANT ANALYSIS 

Discriminant analysis is a procedure for identifying whether 
values on various predictor variables are related to values on 
a grouped criterion variable. The results present a tabulation 
of the object's actual group membership versus their predicted 
group membership [Ref. 17]. In order to predict the probability 
of membership of each individual observation in one of the 
criterion groups, discriminant analysis develops a model using 
the predictor variables shown to have high correlation with the 
criterion variables. Probability of group membership is 
assigned based on the model . Individual observations are 
assigned to the group for which they have the highest probability 

Optionally, discriminant analysis uses a prior probability 
of group membership when assigning predicted group membership. 



22 



(Discriminant Analysis offers the option of assigning either 
actual or equal values to the prior probabilities of member- 
ship in the criterion categories.) Actual probability is 
obtained by running a frequency distribution on the sample 
population. Prior knowledge of group membership increases 
the chance of the discriminant analysis procedure correctly 
assigning individuals into categories based on new predictor 
variables. This study uses the actual proportions of success 
and failure of the sample group. This is felt to be appropri- 
ate since this study is trying to improve on the current 
selection process, and it is realized that all individuals 
have been screened at the time of their enlistment and were 
selected based on their meeting the requirements. 



23 



V. MODELS 

Two separate models were created for those personnel 
assigned to the AX and AW ratings. A general discussion of 
model development for both models will be given followed by a 
separate in-depth discussion of each model . 

From each data base process, two subsets, Deriv8 and 
Valid8, were developed through random sampling for each rating. 
For each rating, Deriv8 was used strictly for developing 
predictor models, and Valid8 was used for validating the 
models . 

A frequency analysis of group membership in the success 
and failure categories was conducted on both ratings to 
determine how well Navy's current assignment process was 
operating. For the AX rating the success rate was 62%, and for 
the AW rating the success rate was 68.5%. The models developed 
by this study would have to better these percentages in order 
to serve as part of an improved assignment process. 

In computing the actual models, two basic statistical 
procedures, stepwise regression and discriminant analysis, 
were utilized. 

A. AX MODEL 

The stepwise regression initially identified four variables 
that best explained the differences between the success and 
failure categories: Term of Enlistment, SCREEN, ASVABNO, and 

24 



ASVABGI . Of the four variables Term of Enlistment had the 
highest R 2 = .1963, meaning that it explained 19.63% of the 
difference between the two categories (see Table 40). After 
careful consideration, the authors chose to delete Term of 
Enlistment as a predictor variable due to the fact that 187 
of the 257 observations had initial enlistments for six years 
and were given automatic advancement to E-4 upon completion 
of Class A School (see Table 39). Based on these facts, 
a large number of observations would fall into the success 
category on the basis of their enlistment contract. Addition- 
ally, Term of Enlistment, used in the strict sense of the word, 
cannot be considered a personal attribute, and is best described 
as an enlistment choice. The decision processes behind offer- 
ing four or six year enlistments were not researched. 

After Term of Enlistment was deleted from the predictor 
variables, stepwise regression then selected the following four 
significant predictor variables: SCREEN, ASVABGI, Entry 
Paygrade, and ASVABNO (see Table 41). No excessively high 
correlations among the four variables were observed. Multi- 
collinearity was not deemed to be an issue. 

The next step involved running a discriminant analysis on 
the second set of predictor variables listed above using prior 
probabilities of 62% and 38%. The results are shown in 
Table 42. 



25 



The positions as shown in the matrix are as follows: 

1. (1,1) The number and percentage of succcessful 
individuals correctly assigned to the successful 
category. 

2. (1,2) The number and percentage of individuals 
assigned to the unsuccessful category who were actual 
successes - "false negatives". 

3. (2,1) The number and percentage of unsuccessful 
individuals incorrectly classified as successful - 
"false positives". 

4. (2,2) The number and percentage of failures correctly 
classified. 

The success of the model can be described by its "hit 
rate". The total hit rate is the percentage of correct 
classifications divided by the total number of classifications 
made. The results produced a hit rate of 66% for the model 
derivation run and 65% for the validation run. 

The results indicate that the model would correctly assign 
4% more individuals than the Navy's current assignment process 
The model incorrectly classified 72.92% of the unsuccessful 
individuals as successes. 

B. AW MODEL 

Of the eighteen variables chosen for analysis, the step- 
wise regression initially identified six predictor variables: 
Term of Enlistment, SCREEN, ASVABAR, ASVABSP, ASVABSI, and 
ASVABGS (see Table 43). For the reasons mentioned in the 
foregoing section, Term of Enlistment was deleted. The 
subsequent stepwise regression yielded the following four 
predictor variables: SCREEN, ASVABAR, ASVABMK, and Entry 



26 



Paygrade. There were no significantly high sample correlations 
between the variables, thus multicollinearity was again not an 
issue. The results are shown in Table 44. 

The model produced a hit rate of 69% (Table 45). When 
compared to Navy's current success rate of 68.5%, negligible 
improvement was attained. This model incorrectly classified 
99% of the unsuccessful individuals as successes. 

C. ADDENDUM 

As a matter of interest, the following results of using 
Term of Enlistment as a predictor variable for the two models 
are provided for possible use in future analysis. 

AX MODEL WITH TERM OF ENLISTMENT, SCREEN, ASVABNO and ASVABGI 
Hit rate: Model 76% Validation 75% 
(correctly assigned failures 69.58% of the time) 
(Table 46) 

AW MODEL WITH TERM OF ENLISTMENT, SCREEN, ASVABSP, ASVABAR, 
ASVABSI, and ASVABGS 
Hit rate: Model 75% Validation 73% 
(correctly assigned failures 64.88% of the time) 
(Table 47) 

The hit rates and failure classification rates appear 
attractive as the hit rates are 13% and 6.5% higher for the 
AX and AW ratings, respectively, than the Navy's. It is emphasized 



27 



that the authors are of the opinion that unless the effects 
of six year enlistments and automatic advancements to E-4 
are controlled for, the results are not useful. 



28 



VI. CONCLUSIONS 

The results obtained from both AX and AW models, when Term 
of Enlistment is not considered, offer a certain amount of 
improvement over the Navy's current assignment process. In 
the case of the AX model developed in this analysis, a 4% 
increase over the Navy's assignment process would translate 
into substantial savings. To a lesser degree the same would 
be true for the .5% increase with the AW model. Of concern 
though is the false success assignment rate produced by both 
models. If the benefits in terms of cost and utility are 
higher by correctly assigning individuals into the AW and AX 
ratings than they are to incorrectly assigning them, then this 
analysis might lend support to modify current AX and AW 
assignment standards. Further study in the areas of cost and 
utility analysis is recommended. Such an analysis should also 
consider the costs and utilities of correct rejections and 
wrong rejections. 

The benefit of this analysis is that given the information 
at the time of enlistment and the definition of success used 
in this study, it was shown that an improvement can be made 
to the AW and AX assignment process. The AX model used ASVABGI 
and ASVABNO vice those currently used by the Navy (MK, EI, GS 
and AR) . A suggested follow on study would be to analyze the 
effects of using different combinations of the ASVA3 subtests. 



29 



The AW model lent support to the Navy's assignment process in 
that it used two of the same variables, i.e., ASVABAR and 
ASVABMK, the Navy currently uses (AR, MK and GS ) . 

The role of Term of Enlistment in predicting success in 
the assignment process deserves further analysis. A suggested 
method would be to separate those individuals with different 
enlistment obligations and run an analysis similar to the one 
used in this study to see how, or if, the people who enlist 
for different lengths of service differ in variables predictive 
of success in the Navy. 

As noted by Whitmire and Deitchman [Ref. 18], the data base 
available for this analysis did not include those individuals 
who were rejected in the current assignment process. Therefore, 
we do not know the Navy's current wrong-rejection rate. Only 
those personnel who were actually assigned to the rating were 
available for analysis. This leaves open the possibility that 
more accurate screening tools could have been used initially. 
And, had those rejected been available, the results of this 
analysis may have been different. 



30 



TABLE 1 
INTER-SERVICE SEPARATION CODE FOR THE AW RATING 



C3 


FREQUENCY 


CUM FREQ 


PERCENT 


CUM PERC 





39 5 


395 


36. 106 


36. 106 


1 


52 6 


921 


48.030 


84. 186 


2 


1 


922 


0.091 


84.278 


8 


14 


936 


1.280 


85.558 


10 


13 


949 


1. 138 


86. 746 


1 1 


4 


953 


0.366 


87. 112 


13 


5 


958 


0.457 


87.569 


16 


1 


959 


0.391 


87.660 


32 


7 


9 66 


0.640 


88.300 


40 


15 


981 


1.371 


89.671 


50 


1 


982 


0.091 


89.762 


60 


13 


9 95 


1. 183 


90.951 


61 


1 


9 96 


0.091 


91. 042 


63 


2 


9 98 


0. 183 


91. 225 


64 


4 


1002 


0.366 


91.590 


65 


23 


1025 


2.102 


93.693 


67 


3 


1023 


0.274 


93.967 


71 


2 


1030 


0. 183 


94. 150 


73 


6 


1036 


0.548 


94.698 


74 


2 


1038 


0. 183 


94. 881 


75 


1 


1039 


0.091 


94. 973 


76 


2 


1041 


0.183 


95. 155 


78 


7 


1048 


0.640 


95.795 


80 


1 


1049 


0.091 


95. 887 


82 


3 


1052 


0.274 


96. 161 


86 


12 


1064 


1.097 


97. 258 


87 


1 


1065 


0.091 


97.349 


91 


15 


1080 


1.371 


98.720 


95 


1 


1081 


0.091 


98.312 


96 


1 


1082 


0.091 


93. 903 


98 


7 


1089 


0.640 


99.54 3 


99 


5 


1094 


0.457 


100.000 



31 



TABLE 2 
INTER-SERVICE SEPARATION CODE FOR AX RATING 



ISC3 


FREQUENCY 


CUM FR2Q 


PERCENT 


CUM PERCENT 





257 


257 


45.975 


45.975 


1 


237 


494 


42. 397 


88.372 


2 


1 


495 


0. 179 


83.551 


8 


10 


50 5 


1.739 


90.340 


10 


3 


508 


0.537 


90.877 


11 


2 


510 


0.358 


91.234 


13 


2 


512 


0.358 


91.592 


22 


6 


518 


1.073 


92.665 


32 


4 


522 


0.716 


93.381 


40 


1 


523 


0. 179 


93.560 


60 


8 


531 


1.431 


94.991 


61 


1 


532 


0. 179 


95.170 


63 


1 


533 


0. 179 


95.349 


64 


1 


53 4 


0. 179 


95.523 


65 


6 


540 


1.073 


96.60 1 


67 


1 


541 


0. 179 


96.780 


71 


1 


542 


0. 179 


96.959 


73 


3 


545 


0.537 


97.496 


76 


2 


547 


0.358 


97.353 


78 


2 


549 


0. 358 


98.21 1 


82 


5 


554 


0.894 


99. 106 


86 


1 


555 


0. 179 


99.284 


90 


1 


556 


0. 179 


99.463 


91 


1 


557 


0. 179 


99.642 


9 


2 


559 


0.358 


100.000 



32 



TABLE 3 
PREDICTOR VARIABLES 



VARIABLE 
ASVABGI (General lat 
ASVABNO (Numerical 
ASVABAD (Attention t 
A3VABWK (Word Knowle 
A37ABAR (Arithmetic 
ASVABSP (Spatial Psr 
A5VABMK (Mathematioa 
A5VABEI (Electronics 
ASVABMC (Mechanical 
ASVABGS (General Sci 
ASVABSI (Shop Inform 
ASVABAI (Automative 
SCREEN (Success Chan 

Recruits Ente 
RACE (1 = Caucasian, 2 
ENTRY PAYGRADE (E1-3 
MARITAL DEPENDENTS ( 

dependents an 
AFQT PERCENTILE (Bas 

ASVAB subtest 
TERM OF ENLISTMENT ( 

of years serv 





AM table 


AX table 


elligence) 


4 


22 


per at ions) 


5 


23 


o Detail) 


6 


24 


dge) 


7 


25 


Reasoning) 


8 


26 


ception) 


9 


27 


1 Knowledge) 


10 


28 


Intelligence) 


11 


29 


Comprehension) 


12 


30 


ence) 


13 


31 


a tion) 


14 


32 


Information) 


15 


33 


ces for 


16 


34 


ring the Navy) 






= Black, 3=3ther) 


17 


35 


) 


18 


36 


# of 






d marital status) 


19 


37 


ed on 






s WK,AR,SP) 


20 


38 


N urn b e r 






ice) 


21 


39 



33 



TABLE 4 
AW ASVAB APTITUDE AREA SCORE — SUBSCALE GI 



ASVABGI 


FREQUENCY 


CUM FREQ 


PERCENT 


CUM PERCENT 





2 


2 


0. 133 


0. 133 


2 


1 


3 


0.091 


0.274 


4 


3 


5 


0.274 


0.548 


5 


3 


9 


0.274 


0.323 


6 


13 


22 


1. 138 


2.011 


7 


22 


44 


2.01 1 


4.022 


8 


50 


94 


4.570 


8.592 


9 


75 


169 


6.856 


15.448 


10 


113 


282 


10.329 


25.777 


1 1 


15 


442 


14.625 


40.402 


12 


201 


643 


18.373 


53.775 


13 


216 


859 


19.744 


78.519 


14 


176 


1035 


16.083 


94.607 


15 


59 


1094 


5.393 


1 00.000 



34 



TABLE 5 
AW ASVAB APTITUDE AREA SCORE — SUBSCALE NO 



3VABN0 


FREQUENCY 


CUM FREQ 


PERCENT 


CUM PERCENT 





2 


2 


0.183 


0. 183 


6 


2 


4 


0.183 


0.366 


9 


1 


5 


0.091 


0.457 


13 


2 


7 


0.133 


0.640 


14 


2 


9 


0.183 


0.823 


15 


3 


12 


0.274 


1 .097 


16 


1 


13 


0.091 


1. 188 


17 


3 


16 


0.274 


1.463 


18 


1 


17 


0.091 


1.554 


19 


9 


26 


0.823 


2.377 


20 


9 


35 


0.323 


3. 199 


21 


11 


46 


1 .005 


4.205 


22 


12 


58 


1.097 


5. 302 


23 


13 


71 


1.188 


6.490 


24 


19 


90 


1.737 


8.227 


25 


13 


1 03 


1.188 


9.415 


26 


24 


1 27 


2.194 


11.609 


27 


29 


156 


2.651 


14.260 


28 


30 


1 86 


2.742 


17.002 


29 


37 


2 23 


3.382 


20.384 


30 


55 


278 


5.027 


25.411 


31 


53 


3 36 


5.302 


30.713 


32 


55 


391 


5.027 


35.740 


33 


55 


446 


5.027 


40.768 


34 


64 


5 10 


5.350 


46.618 


35 


62 


572 


5.667 


52.285 


36 


44 


5 16 


4.022 


56.307 


37 


60 


6 76 


5.484 


61.792 


38 


49 


7 25 


4.479 


6 6. 271 


39 


44 


769 


4.022 


70.293 


40 


49 


3 18 


4.479 


74.771 


41 


25 


843 


2.285 


77.057 


42 


42 


8 85 


3.839 


30.896 


43 


41 


9 26 


3.748 


34.644 


44 


19 


9 45 


1.737 


36.380 


45 


23 


968 


2.102 


38.483 


46 


31 


9 99 


2.834 


91.316 


47 


21 


10 20 


1 .920 


93.236 


48 


20 


1040 


1.828 


95.064 


49 


22 


1062 


2.011 


97.075 


50 


32 


1094 


2.925 


100.000 



35 



TABLE 6 
AW ASVAB APTITUDE AREA SCORE — SUBSCALE AD 



CUM FREQ PERCENT CUM PERCENT 

2 0.183 0.183 

3 0.091 0.274 
5 0.183 0.457 
8 0.274 0.731 

17 0.823 1.554 

29 1.3 97 2.651 

54 2.285 4.936 

98 4.022 8.958 

154 5.119 14.077 

251 8.867 22.943 

376 11.426 34.369 

i* 92 10.503 44.973 

630 12.614 57.587 

741 10.146 67.733 

843 9.324 77.057 

935 3.410 85.466 

992 5.210 90.676 

1023 2.834 93.510 

1055 2.925 96.435 

1073 1.645 98.080 

1081 0.731 98.812 

1088 0.640 99.452 

1091 0.274 99.726 

1093 0.183 99.909 

1094 0.091 100.000 



ASVABAD 


FREQUENCY 





2 


3 


1 


5 


2 


6 


3 


7 


9 


8 


12 


9 


25 


10 


44 


11 


56 


12 


97 


13 


125 


14 


116 


15 


138 


16 


1 11 


17 


102 


18 


92 


19 


57 


20 


31 


21 


32 


22 


18 


23 


8 


24 


7 


25 


3 


26 


2 


28 


1 



36 



TABLE 7 
AW ASVAB APTITUDE AREA SCORE — SUBSCALE WK 



ASVABWK 


FREQUENCY 


CUM FREQ 


PERCENT 


CUM PERCENT 





2 


2 


0.183 


0.183 


6 


1 


3 


0.091 


0. 274 


9 


1 


4 


0.091 


0.366 


13 


2 


6 


0.183 


0.548 


14 


2 


8 


0.183 


0.731 


15 


4 


12 


0.366 


1.097 


16 


11 


23 


1.005 


2.102 


17 


13 


36 


1.188 


3.291 


18 


18 


54 


1 .645 


4.936 


19 


33 


87 


3.016 


7.952 


20 


48 


1 35 


4.388 


12.340 


21 


47 


1 82 


4.296 


16.6 36 


22 


60 


2 42 


5.484 


22.121 


23 


80 


3 22 


7.313 


29.433 


24 


103 


4 25 


9.415 


38.848 


25 


94 


5 19 


8.592 


47.441 


26 


1 19 


638 


10.878 


58.318 


27 


129 


767 


1 1.792 


70. 1 10 


28 


1 14 


881 


10.420 


80.530 


29 


116 


997 


10.603 


91. 133 


30 


97 


1094 


8.867 


100.000 



37 



TABLE 8 
AW ASVAB APTITUDE AREA SCORE — SUBSCALE AR 



CUM FREQ PERCENT CUM PERCENT 

2 D. 183 0.183 

4 0.183 0.366 

5 0.09 1 0.457 
8 0.274 0.731 

17 0.323 1.554 

41 2.194 3.74 8 

77 3.291 7.038 

131 4.936 11.974 

207 5.947 13.921 

325 10.786 29.707 

472 13.437 43.144 

632 14.625 57.770 

759 12.523 70.293 

905 12.431 82.724 

1017 10.238 92.962 

1094 7.038 100.000 



ASVABAR 


FREQUENCY 





2 


2 


2 


7 


1 


8 


3 


9 


9 


10 


24 


11 


36 


12 


54 


13 


76 


14 


1 18 


15 


147 


16 


160 


17 


137 


18 


136 


19 


112 


20 


77 



38 



TABLE 9 
AW ASVAB APTITUDE AREA SCORE — SUBSCALE SP 



FREQ PERCENT CUM 



A5VABSP 


FREQUENCY 


CUM 





3 


3 


3 


2 


5 


4 


4 


9 


5 


5 


14 


6 


9 


23 


7 


19 


42 


8 


46 


88 


9 


52 


140 


10 


53 


193 


11 


93 


286 


12 


104 


39 


13 


103 


493 


14 


93 


586 


15 


1 18 


704 


16 


106 


810 


17 


107 


917 


18 


66 


983 


19 


66 


1049 


20 


45 


1094 



P5 


IRCENT 


0. 


274 


0. 


183 


0. 


366 


3. 


457 


0. 


823 


1. 


737 


4. 


205 


4. 


753 


4. 


845 


3. 


501 


9. 


506 


9. 


415 


3. 


501 


10. 


786 


9. 


689 


9. 


781 


6. 


033 


6. 


033 


4. 


113 



M PERCENT 


0. 


274 


0. 


457 


0. 


,823 


1. 


280 


2. 


,102 


3. 


839 


8. 


,044 


12. 


797 


17. 


,642 


26. 


143 


35. 


,649 


45. 


064 


53. 


,565 


54. 


351 


74. 


,040 


83. 


821 


89. 


.854 


95. 


,887 


100. 


,000 



39 



TABLE 10 
AW ASVAB APTITUDE AREA SCORE — SUBSCALE MK 



ASVA3MK 


FREQUENCY 


CUM FREQ 


PERCENT 


CUM PERCENT 





2 


2 


0. 183 


0. 183 


3 


1 


3 


0.091 


0. 274 


4 


1 


4 


0.091 


0.366 


5 


3 


7 


D.274 


0.640 


6 


14 


21 


1.280 


1.920 


7 


15 


36 


1. 371 


3. 291 


8 


28 


64 


2. 559 


5.350 


9 


41 


105 


3. 748 


9.598 


10 


50 


155 


4.570 


14. 168 


11 


70 


225 


6.399 


20.567 


12 


88 


313 


3.044 


28.611 


13 


1 12 


425 


10.233 


38.848 


14 


106 


531 


9.689 


48.537 


15 


102 


633 


9.324 


57.861 


16 


103 


736 


9.415 


67.276 


17 


93 


829 


3.501 


75.777 


18 


98 


927 


3.958 


84.735 


19 


88 


1015 


3.044 


92.779 


20 


79 


1094 


7.221 


100.000 



40 



TABLE 11 
AW ASVAB APTITUDE AREA SCORE — SUBSCALE EI 



SVABEI 


FREQUENCY 


CUM FSEQ 


P2RCSNT 


CUM PERCENT 





2 


2 


0.183 


0.183 


6 


2 


4 


0.183 


0. 366 


7 


2 


6 


0.183 


0.548 


9 


4 


10 


0.366 


0.914 


10 


4 


14 


0.366 


1.280 


11 


3 


17 


3.274 


1.554 


12 


7 


24 


0.640 


2. 194 


13 


13 


37 


1. 188 


3.382 


14 


19 


56 


1.737 


5. 119 


15 


25 


81 


2.285 


7.404 


16 


37 


1 18 


3.382 


10.786 


17 


56 


1 74 


5. 1 19 


15.905 


18 


55 


2 29 


5.027 


20.932 


19 


71 


300 


6.490 


27.422 


20 


88 


3 88 


3.044 


35.466 


21 


101 


4 89 


9.232 


44.698 


22 


102 


591 


9.324 


54.022 


23 


1 15 


7 06 


10.512 


64. 534 


24 


99 


805 


9.049 


73.583 


25 


39 


8 94 


8.135 


81.718 


26 


72 


966 


6.581 


88.300 


27 


55 


1021 


5.027 


93.327 


28 


38 


1059 


3.473 


96.801 


29 


22 


1081 


2.011 


98.812 


30 


13 


1094 


1.188 


100.000 



41 



TABLE 12 
AW ASVAB APTITUDE AREA SCORE — SUBSCALE MC 



CUM FREQ PERCENT CUM PERCENT 

2 0.183 0.183 

6 0.366 0.548 

12 0.548 1.097 

28 1.463 2.559 

63 3.199 5.759 

97 3.108 8.867 

154 5.210 14.077 

241 7.952 22.029 

333 8.410 30.439 

4 34 9.232 39.671 

559 11.426 51.097 

6 87 11.700 62.797 

794 9.781 72.578 

902 9.872 82.450 

980 7.130 89.580 

1051 5.490 96.069 

1084 3.016 99.086 

1094 0.914 100.000 



ASVABMC 


FREQUENCY 





2 


4 


4 


5 


6 


6 


16 


7 


35 


8 


34 


9 


57 


10 


87 


1 1 


92 


12 


101 


13 


125 


14 


128 


15 


107 


16 


108 


17 


78 


18 


71 


19 


33 


20 


10 



42 



TABLE 13 
AW ASVAB APTITUDE AREA SCORE — SUBSCALE GS 



CUM FREQ PERCENT COM PERCENT 

2 0.183 0.183 

4 0.183 0.366 

6 0.183 0.548 

15 0.823 1.371 

33 1.645 3.016 

51 1.645 4.662 

86 3.199 7.861 

150 5.350 13.711 

243 8.501 22.212 

338 8.684 30.896 

468 11.883 42.779 

604 12.431 55.210 

723 10.878 66.088 

840 10.695 76.782 

947 9.781 86.563 

1036 8.135 94.698 

1075 3.565 98.263 

1094 1.737 100.000 



ASVABGS 


FREQUENCY 





2 


2 


2 


4 


2 


6 


9 


7 


18 


8 


18 


9 


35 


10 


64 


11 


93 


12 


95 


13 


130 


14 


136 


15 


1 19 


16 


1 17 


17 


107 


18 


89 


19 


39 


20 


19 



43 



TABLE 14 
AW ASVAB APTITUDE AREA SCORE — SUBSCALE SI 



CUM FREQ PERCENT CUM PERCENT 

16 1.463 1.463 

18 0.183 1.645 

19 0.091 1.737 
23 0.366 2.102 
32 0.823 2.925 
42 0.914 3.839 
58 1.463 5.302 

107 4.479 9.781 

146 3.565 13.346 

201 5.027 18.373 

299 8.958 27.331 

407 9.872 37.203 

518 10.146 47.349 

655 12.523 59.872 

794 12.706 72.578 

917 11.243 83.821 

1031 10.420 94.241 

1094 5.759 100.000 



ASVABSI 


FREQUENCY 





16 


2 


2 


5 


1 


6 


4 


7 


9 


8 


10 


9 


16 


10 


49 


1 1 


39 


12 


55 


13 


98 


14 


108 


15 


1 11 


16 


137 


17 


139 


18 


123 


19 


1 14 


20 


63 



44 



TABLE 15 
AW ASVAB APTITUDE AREA SCORE — SUBSCALE AI 



VABAI 


FREQUENCY 


CUM FREQ 


PERCENT 


CUM PERCENT 





17 


17 


1.554 


1.554 


2 


1 


18 


0.091 


1.645 


3 


2 


20 


0-183 


1.828 


4 


10 


30 


0.914 


2.742 


5 


5 


35 


0.457 


3. 199 


6 


19 


54 


1.737 


4.936 


7 


38 


92 


3.473 


8.410 


8 


66 


1 58 


6.033 


14.442 


9 


54 


2 12 


4.936 


19. 378 


10 


71 


283 


6.490 


25.868 


11 


89 


372 


8.135 


34.004 


12 


100 


4 72 


9.141 


43.144 


13 


97 


569 


8.867 


52.011 


14 


90 


659 


8.227 


60.238 


15 


88 


747 


8.044 


68.282 


16 


89 


8 36 


8.135 


76.417 


17 


60 


8 96 


5.484 


81.901 


18 


81 


9 77 


7.404 


89.305 


19 


77 


10 54 


7.038 


96. 344 


20 


40 


1094 


3.656 


100.000 



45 







TABLE 16 






Att 


r SCREEN SCORE 


SCREEN 


FREQUENCY 


CUM FREQ 


PERCENT 


• 


43 


• 


• 


66 


6 


6 


0.571 


68 


2 


8 


0.190 


70 


3 


16 


0.761 


72 


8 


24 


0.751 


74 


23 


52 


2.664 


76 


9 


61 


0.856 


77 


22 


83 


2.093 


78 


35 


118 


3.330 


79 


60 


178 


5.709 


80 


1 


179 


0.095 


81 


39 


218 


3.711 


82 


71 


289 


6.755 


83 


18 


307 


1.713 


84 


40 


34 7 


3.306 


86 


25 


373 


2.474 


87 


93 


466 


8.849 


38 


144 


610 


13.701 


89 


49 


659 


4.652 


90 


323 


982 


30.733 


91 


2 


984 


0.190 


92 


17 


100 1 


1.618 


93 


13 


1014 


1.237 


94 


2 


1016 


0.190 


95 


31 


1047 


2.950 


96 


% 


1051 


0.381 



CUM PERCENT 



0.571 
0.761 

1.522 
2.284 
4.948 
5.804 
7.897 

1 1.227 
16.936 
17.031 
20.742 

2 7.49 8 
29.210 
33.016 
35.490 
44.339 
5 8.04 
62.702 
93.435 
93.625 
95.243 
96.480 
96.670 
99.6 19 

100.000 



46 



TABLE 17 
AW RACE DISTRIBUTION 

(1) WHITE, (2) BLACK, (3) OTHER 
RACE FREQUENCY CUM FREQ PERCENT CUM PERCENT 

1 1048 1048 95.795 95.795 

2 38 1086 3.473 99.269 

3 8 1094 0.731 100.000 



TABLE 18 

AW ENTRY PAY GRADE (E00-011) 

ENTRPAYG FREQUENCY CUM FREQ PERCENT CUM PERCENT 

1 812 812 74.223 74.223 

2 151 963 13.803 88.026 

3 131 1094 11.974 100.000 



TABLE 19 
AW MARITAL STATUS/DEPENDENTS 



MRTLDPND 


FREQUENCY 


CUM FREQ 


PERCENT 


CUM PERCENT 


10 


1051 


1051 


96.069 


96.069 


11 


4 


1055 


0.366 


96.435 


12 


2 


10 57 


0.183 


96.618 


21 


24 


1081 


2.194 


98.812 


22 


12 


1093 


1 .097 


99.909 


24 


1 


10 94 


0.091 


100.000 



47 







TABLE 20 








AW AFQT SCORE FR! 


EQUENCY 




AFQTFCNT 


FREQUENCY CUM FREQ 


PERCENT 


CUM PER 





2 


2 


0. 133 


0. 183 


12 


1 


3 


0.091 


0.274 


17 


1 


4 


0.091 


0.366 


19 


1 


5 


0.091 


0.4 57 


23 


2 


7 


0.133 


0.640 


27 


2 


9 


0.183 


0.823 


29 


1 


10 


0.091 


0.914 


31 


3 


13 


0.274 


1.188 


33 


3 


16 


0.274 


1.463 


35 


9 


25 


0.823 


2.285 


38 


6 


31 


0.548 


2.834 


41 


11 


42 


1.005 


3.839 


44 


17 


59 


1.554 


5.393 


47 


16 


75 


1.463 


6.856 


50 


23 


98 


2.102 


8.958 


53 


29 


127 


2.651 


1 1.609 


56 


42 


169 


3.839 


15.44 8 


58 


50 


219 


4.570 


20.018 


60 


58 


277 


5.302 


25.320 


62 


65 


34 2 


5.941 


31.261 


65 


58 


400 


5.302 


36.563 


67 


79 


479 


7.221 


43.784 


70 


63 


547 


6.216 


50.000 


72 


51 


59 8 


4.662 


54.662 


75 


68 


666 


6.216 


60.878 


77 


51 


717 


4.662 


65.539 


80 


53 


770 


4.345 


70.384 


82 


66 


336 


6.033 


76.417 


84 


46 


882 


4.205 


80.622 


86 


44 


926 


4.022 


84.644 


87 


34 


96 


3.108 


87.751 


89 


34 


994 


3.108 


90.859 


91 


20 


1014 


1.328 


92.637 


93 


23 


1037 


2.102 


94.790 


95 


24 


106 1 


2.194 


96.984 


97 


13 


1074 


1.188 


98.172 


98 


9 


1083 


0.823 


98.995 


99 


11 


1094 


1.005 


100.000 



48 



TABLE 21 
AW TERM OF ENLISTMENT (NO. OF YEARS) 

TERMENLT FREQUENCY CUM FREQ PERCENT CUM PERCENT 

2 4 4 0.366 0.366 

4 657 661 60.055 60.420 

6 433 1094 39.580 100.000 



TABLE 22 
AX ASVAB APTITUDE AREA SCORE — SUBSCALE GI 



ASVABGI 


FREQUENCY 


CUM FREQ PERCENT 


CUM PERCENT 


2 


2 


2 


0.358 


0.358 


5 


4 


6 


0.716 


1.073 


6 


8 


14 


1.-4 3 1 


2.504 


7 


8 


22 


1.431 


3.936 


8 


18 


40 


3.220 


7. 156 


9 


42 


82 


7.513 


14.669 


10 


63 


145 


1 1.270 


25.939 


11 


57 


212 


11.986 


37.925 


12 


113 


325 


20.215 


58. 140 


13 


110 


435 


19.678 


77.818 


14 


35 


520 


15.206 


93.0 23 


15 


39 


559 


6.977 


100.000 



49 



TABLE 23 
AX ASVAB APTITUDE AREA SCORE — 3UBSCALE NO 



SVABNO 


FREQUENCY 


CUM FREQ 


PERCENT 


CUM PERC 


7 


1 


1 


0. 179 


0. 179 


13 


1 


2 


0. 179 


0.358 


15 


1 


3 


0. 179 


0.53 7 


16 


1 


4 


3. 179 


0.716 


17 


2 


6 


0. 358 


1.073 


18 


3 


9 


0.537 


1.610 


19 


1 


10 


0. 179 


1 .789 


20 


4 


14 


3. 716 


2.504 


21 


5 


1 9 


0.894 


3.399 


22 


5 


24 


3.894 


4.293 


23 


9 


33 


1. 610 


5.903 


24 


8 


4 1 


1. 431 


7.335 


25 


12 


53 


2. 147 


9.481 


26 


12 


65 


2. 147 


11.628 


27 


22 


87 


3.936 


15.564 


28 


18 


105 


3. 220 


18.784 


29 


21 


126 


3.757 


22.540 


30 


22 


148 


3.936 


26.476 


31 


13 


16 1 


2. 326 


28.801 


32 


38 


199 


5.798 


35.599 


33 


22 


22 1 


3.936 


39.535 


34 


30 


25 1 


5. 367 


44.902 


35 


17 


26 8 


3.041 


47.943 


36 


29 


297 


5. 138 


53. 131 


37 


30 


327 


5.367 


58.497 


38 


30 


357 


5. 367 


63. 364 


39 


29 


336 


5. 188 


69.C52 


40 


25 


41 1 


4.47 2 


73.524 


41 


26 


437 


4.651 


78.175 


42 


19 


456 


3.399 


81.574 


43 


16 


472 


2. 862 


84.436 


44 


15 


487 


2. 633 


87. 120 


45 


1 1 


498 


1 .968 


89.088 


46 


12 


510 


2. 147 


91.234 


47 


12 


522 


2. 147 


93.381 


48 


9 


53 1 


1.610 


94.991 


49 


12 


54 3 


2. 147 


97. 138 


50 


16 


559 


2. 852 


100.000 



50 



TABLE 24 
AX ASVAB APTITUDE AREA SCORE — SUBSCALE AD 



1SVABAD 


FREQUENCY 


COM FREQ PERCENT 


CUM PERCENT 


3 


1 


1 


0. 179 


0. 179 


a 


1 


2 


0.179 


0.358 


5 


1 


3 


0.179 


0.537 


6 


1 


4 


0.179 


0.716 


7 


2 


5 


0.358 


1.073 


8 


1 1 


17 


1.968 


3.041 


9 


13 


30 


2.326 


5. 367 


10 


22 


52 


3.936 


9.302 


11 


31 


83 


5.546 


14. 848 


12 


50 


133 


8.945 


23.792 


13 


49 


182 


8.756 


32.558 


14 


70 


252 


12.522 


45.081 


15 


60 


312 


10.733 


55.814 


16 


66 


378 


11.807 


67.621 


17 


43 


421 


7.592 


75.313 


18 


41 


462 


7.335 


82.648 


19 


32 


4 94 


5.725 


88. 372 


20 


19 


513 


3.399 


91.771 


21 


17 


530 


3.041 


94.812 


22 


1 1 


541 


1.968 


96.730 


23 


6 


547 


1.373 


97.853 


24 


6 


553 


1.073 


98.927 


25 


3 


556 


0.537 


99.463 


26 


3 


559 


0.537 


100.000 



51 



TABLE 2 5 
AX ASVAB APTITUDE AREA SCORE — SUBSCALE WK 



ASVA3WK 


FREQUENCY 


CUM FREQ 


PERCENT 


CUM PERCENT 


5 


1 


1 


0.179 


0.179 


8 


1 


2 


0.179 


0.358 


12 


3 


5 


0.537 


0.894 


14 


2 


7 


0.358 


1.252 


15 


3 


10 


0.537 


1.789 


16 


5 


15 


0.894 


2.683 


17 


7 


22 


1.252 


3.936 


18 


12 


34 


2. 147 


6.082 


19 


17 


51 


3.041 


9.123 


20 


18 


69 


3.220 


12.343 


21 


32 


101 


5.725 


18.068 


22 


34 


1 35 


6.082 


24. 150 


23 


31 


166 


5.546 


29.696 


24 


40 


2 06 


7.156 


36.852 


25 


45 


251 


8.050 


44.902 


26 


48 


299 


8.587 


5 3.4 88 


27 


60 


359 


10.733 


64.222 


28 


68 


4 27 


12. 165 


76.386 


29 


63 


490 


1 1.270 


87.657 


30 


69 


559 


12.343 


100.000 



52 



TABLE 26 
AX ASVAB APTITUDE AREA SCORE — SUBSCALE AR 



ASVABAR 


FREQUENCY 


CUM FREQ 


PERCENT 


CUM PERCENT 


6 


1 


1 


3.179 


0. 179 


8 


1 


2 


0.179 


0.358 


9 


3 


5 


0.537 


0.894 


10 


5 


10 


0.894 


1.789 


11 


10 


20 


1.789 


3.578 


12 


23 


43 


4.1 14 


7.692 


13 


26 


69 


4.651 


12.343 


14 


43 


1 12 


7.692 


20.036 


15 


47 


159 


8.408 


28.444 


16 


74 


233 


1 3.238 


41 .682 


17 


103 


3 36 


18.426 


60. 107 


18 


82 


4 18 


14.669 


74.776 


19 


73 


491 


13.059 


87.835 


20 


68 


559 


12.165 


100.000 



53 



TABLE 27 
AX ASVAB APTITUDE AREA SCORE - SUBSCALE SP 



ASVABSP 


FREQUENCY 


COM 


FREQ PERCENT 


CUM PERCENT 


4 


1 


1 


0.179 


0.179 


5 


2 


3 


0.358 


0.537 


6 


5 


8 


0.894 


1.431 


7 


12 


20 


2.147 


3.578 


3 


3 


28 


1.431 


5.009 


9 


19 


47 


3.399 


8.408 


10 


28 


75 


5.009 


13.417 


11 


25 


100 


4.472 


17.839 


12 


34 


134 


6.082 


23.971 


13 


46 


180 


8.229 


32.200 


14 


55 


236 


10.018 


42.218 


15 


54 


290 


9.S50 


5 1.378 


16 


56 


346 


10.018 


61.896 


17 


65 


41 1 


11.623 


7 3.5 24 


18 


63 


474 


11.270 


84.794 


19 


53 


527 


9.431 


94.275 


20 


32 


559 


5.725 


100.000 



54 



TABLE 28 
AX ASVAB APTITUDE AREA SCORE — SUBSCALE MK 



ASVABMK 


FREQUENCY 


COH 


FREQ PERCENT 


CUM PERCENT 


7 


1 


1 


0.179 


0.179 


8 


3 


4 


0.537 


0.716 


9 


9 


13 


1.610 


2.326 


10 


8 


21 


1.431 


3.7 57 


1 1 


22 


43 


3.93 6 


7.692 


12 


18 


61 


3.220 


10.912 


13 


27 


88 


4.830 


15.742 


14 


40 


123 


7.155 


22.898 


15 


63 


196 


12.165 


35.063 


16 


72 


26 8 


12.830 


47.943 


17 


83 


356 


15.742 


63.685 


18 


72 


428 


12.330 


76.565 


19 


77 


50 5 


13.775 


90.340 


20 


54 


55 9 


9.550 


100.000 



55 



TABLE 29 
AX ASVAB APTITUDE AREA SCORE -- SUBSCALE EI 



ASVABEI 


FREQUENCY 


CUM FREQ 


PERCENT 


CUM PERCENT 


11 


1 


1 


0.179 


0. 179 


13 


1 


2 


0.179 


0.358 


14 


2 


4 


0.358 


0.716 


16 


6 


10 


1 .073 


1.789 


17 


5 


15 


0.394 


2.683 


18 


10 


25 


1.789 


4.472 


19 


19 


44 


3.399 


7.871 


20 


33 


77 


5.903 


13.775 


21 


32 


1 09 


5.725 


19.499 


22 


35 


1 44 


6.261 


25.760 


23 


64 


2 08 


1 1.449 


37.209 


24 


44 


2 52 


7.871 


45.081 


25 


61 


313 


10.912 


55.993 


26 


79 


392 


14.132 


70.125 


27 


64 


4 56 


1 1 .449 


81.574 


28 


49 


5 05 


8.766 


90.340 


29 


35 


540 


S.261 


96.601 


30 


19 


5 59 


3.399 


100.000 



56 



TABLE 30 
AX ASVAB APTITUDE AREA SCORE — SUBSCALE MC 



ASVABMC 


FREQUENCY 


CUM FREQ 


PERCENT 


CUM PERCENT 


4 


2 


2 


0.358 


0.353 


5 


1 


3 


0.179 


0.5 37 


7 


6 


9 


1 .073 


1.610 


8 


9 


18 


1.610 


3.220 


9 


13 


31 


2.326 


5.546 


10 


29 


60 


5. 188 


10.733 


1 1 


44 


1 04 


7.871 


18.605 


12 


43 


147 


7.692 


26.297 


13 


59 


2 06 


10.555 


36.852 


14 


62 


268 


1 1 .091 


47.943 


15 


63 


331 


1 1 .270 


59.213 


16 


67 


3 98 


1 1.986 


71. 199 


17 


61 


4 59 


10.912 


82.111 


18 


59 


5 13 


10.555 


92.665 


19 


28 


5 46 


5.009 


97.674 


20 


13 


559 


2.326 


100.000 



57 



TABLE 31 
AX ASVAB APTITUDE AREA SCORE — SUBSCALE GS 

CUM FREQ PERCENT CUM PERCENT 

2 0.358 0.358 

7 0.8 94 1.2 52 

14 1.252 2.504 

34 3.573 6.082 

66 5.725 11.807 

130 11.449 23.256 

200 12.522 35.778 

276 13.596 49.374 

367 16.279 65.653 

450 14.848 80.501 

503 9-.481 89.982 

543 7.156 97.138 

559 2.862 100.000 



ASVABGS 


FREQUENCY 


8 


2 


9 


5 


10 


7 


11 


20 


12 


32 


13 


64 


14 


70 


15 


76 


16 


91 


17 


83 


18 


53 


19 


40 


20 


16 



58 



TABLE 32 
AX ASVAB APTITUDE AREA SCORE — SUBSCALE SI 

CUM FREQ PERCENT CUM PERCENT 

6 1.073 

7 0.179 
3 0.179 

13 0.894 

13 0.894 

34 2.862 

48 2.504 

65 3.041 

91 4.651 

137 8.229 

182 3.050 

247 11.628 

329 14.669 

430 18.068 

511 14.490 

559 8.587 



VABSI 


FREQUENCY 





6 


3 


1 


5 


1 


8 


5 


9 


5 


10 


16 


1 1 


14 


12 


17 


13 


26 


14 


46 


15 


45 


16 


65 


17 


82 


18 


101 


19 


81 


20 


48 



1. 


,073 


1. 


,252 


1, 


,431 


2, 


,326 


3. 


,220 


6. 


.082 


8. 


587 


1 1, 


,628 


16. 


279 


24. 


.508 


32. 


558 


44. 


,186 


58. 


855 


76. 


,923 


91. 


,413 


100. 


,000 



59 



TABLE 3 3 
AX ASVAB APTITUDE AREA SCORE — SUBSCALE AI 



ASVABAI 


FREQUENCY 


CUM FREQ 


PERCENT 


CUM PERCENT 





6 


6 


1.073 


1.073 


4 


2 


8 


0.358 


1.431 


5 


6 


14 


1.073 


2.504 


6 


5 


19 


0.394 


3.399 


7 


8 


27 


1.431 


4.830 


8 


13 


40 


2.326 


7. 156 


9 


23 


63 


4.114 


1 1 .270 


10 


28 


91 


5.009 


16.279 


1 1 


42 


133 


7.513 


23.792 


12 


33 


166 


5.903 


29.696 


13 


41 


2 07 


7.335 


37.030 


14 


51 


258 


9.123 


46. 154 


15 


45 


3 03 


8.050 


54.204 


16 


41 


344 


7.335 


61.538 


17 


64 


408 


11.449 


72.987 


18 


54 


462 


9.660 


82.648 


19 


60 


5 22 


10.733 


93.381 


20 


37 


559 


6.619 


100.000 



60 







TABLE 34 






AX 


SCREEN Si 


CORE 


SCREEN 


FREQUENCY 


CUM FRSQ 


PERCENT 


• 


36 


• 


• 


66 


5 


6 


1.147 


70 


6 


12 


1.147 


72 


2 


14 


0.382 


7U 


6 


20 


1.147 


76 


5 


25 


0.956 


77 


H 


29 


0.755 


78 


21 


50 


4.015 


79 


15 


65 


2.353 


81 


5 


70 


0.956 


82 


55 


125 


10.515 


83 


3 


133 


1.530 


84 


10 


143 


1.912 


86 


23 


166 


4.398 


87 


48 


214 


9.178 


88 


63 


27 4 


1 1.472 


89 


24 


29 8 


4.539 


90 


179 


477 


34.226 


91 


1 


478 


0.191 


92 


13 


48 8 


1.912 


93 


10 


498 


1.912 


94 


1 


499 


0.191 


95 


22 


52 1 


4.237 


96 


2 


523 


0.382 



CUM PERCENT 



1.147 

2.294 

2.677 

3.824 

4.780 

5.545 

9.560 

12.428 

13.384 

23.90 1 

25.430 

27.342 

31.740 

40.918 

52.390 

56.979 

91.205 

91.396 

93.308 

95.220 

95.41 1 

99.618 

100,000 



61 



TABLE 35 
AX RACE DISTRIBUTION 

(1) WHITE, (2) BLACK, (3) OTHER 
RACE FREQUENCY CUM FREQ PERCENT CUM PERCENT 



1 


529 


529 


94.633 


94.633 


2 


20 


549 


3.578 


98.211 


3 


10 


559 


1.739 


100.000 



TABLE 3 6 
AX ENTRY PAY GRADE (E0 0-011) 



ENTRPAYG 


F RE QUE 


NCY 


CUM 


FREQ 


I 


?SRC 


1 
2 
3 
6 


280 

35 

243 

1 




280 
315 
558 
5 59 




50. 

6, 
43. 

0, 


.089 
.261 
.470 
.179 



PERCENT CUM PERCENT 



50.089 

56.351 

99.821 

100.000 



TABLE 3 7 
AX MARITAL STATUS/DEPENDENTS 



MRTLDPND 


FREQUENCY 


COB FREQ 


PERCENT 


CUM PERCENT 


10 


5 20 


5 20 


93.023 


93.023 


11 


5 


5 25 


0.894 


93.918 


12 


4 


5 29 


0.716 


94.633 


14 


1 


5 30 


0.179 


94.812 


20 


1 


531 


0.179 


94.991 


21 


19 


550 


3.399 


98.390 


22 


8 


5 58 


1.431 


99.821 


23 


1 


559 


0.179 


100.000 



62 



TABLE 38 
AX AFQT PERCENTILE (OR EQUIVALENT) 



QTPCNT 


FREQUENCY 


CUM FREQ 


PERCENT 


CUM PERCENT 


17 


1 


1 


0.179 


0. 179 


23 


1 


2 


0.179 


0.358 


25 


1 


3 


0.179 


0.5 37 


27 


1 


4 


0. 179 


0.716 


29 


1 


5 


0.179 


0.894 


31 


2 


7 


0.358 


1.252 


33 


1 


8 


0. 179 


1.431 


35 


1 


9 


0.179 


1.610 


38 


5 


14 


0.894 


2. 504 


41 


5 


19 


0.894 


3.399 


44 


4 


23 


0.716 


4. 114 


47 


4 


27 


0.716 


4.830 


50 


6 


33 


1.073 


5.903 


53 


9 


42 


1 .610 


7.513 


56 


11 


53 


1.968 


9.481 


58 


28 


81 


5.009 


14.490 


60 


19 


1 00 


3.399 


17.889 


62 


18 


1 18 


3.220 


21 .109 


65 


25 


143 


4.472 


25.581 


67 


29 


172 


5.188 


30.769 


70 


33 


2 05 


5.903 


36.673 


72 


32 


237 


5.725 


4 2.3 97 


75 


22 


259 


3.936 


46. 333 


77 


33 


292 


5.903 


52.236 


80 


36 


3 28 


6.440 


58.676 


82 


26 


3 54 


4.651 


63.327 


84 


34 


3 88 


6.082 


69.410 


86 


18 


4 06 


3.220 


72.630 


87 


34 


440 


5.082 


78.712 


89 


32 


472 


5.725 


84.436 


91 


21 


4 93 


3.757 


88. 193 


93 


20 


5 13 


3.578 


91.771 


95 


18 


531 


3.220 


94.991 


97 


14 


545 


2.504 


97.496 


98 


10 


5 55 


1.789 


99.284 


99 


4 


559 


0.716 


100.000 



63 



TABLE 39 

AX TERM OF ENLISTMENT (NO. OF YEARS) 

TERMSNLT FREQUENCY CUM FRE Q PERCENT COM PERCENT 

2 1 1 0.179 0.179 

4 371 372 66.369 66.547 

6 187 559 33.453 100.000 



TABLE 40 

AX STEPWISE SELECTION: SUMMARY 
TERMENLT AS A VARIABLE 



Variable 
Entered 


Number 
In 


Partial 
R**2 


F 
StatF 


Prob 


TERMSNLT 
SCREEN 
ASVABNO 
ASVABGI 


1 
2 
3 
4 


0.1963 
0.0229 
0.0169 

0.0095 


6 1.5 33 
5.878 
4.289 
2.395 


3.0001 
0.0160 
0.03 94 
0. 1230 



TABLE 41 

AX STEPWISE SELECTION: SUMMARY 
WITHOUT TERMENLT 



VARIABLES 

Entered 


Number 
In 


Partial 
R**2 


F 
Stat 


Prob 
F 


SCREEN 
ASVABGI 
ENTRPAYG 
ASVSABNO 


1 

2 

3 
4 


0.0304 
0.0237 
0.0139 
0.0090 


7.898 
5.303 
3.526 
2.266 


0.0053 
0.0221 
0.0616 
0. 1335 



64 



TABLE 42 
AX DISCRIMINANT ANALYSIS 

Deriv8 WITHOUT TERMENLT 

From 

C1 1 2 Toi:al 

12 12 

130.0 0.0 100.0 

1 156 13 159 
92.31 7.69 100.0 

2 76 29 105 
72.38 27.62 100.0 

Total 244 42 235 

Percent 85.31 14.69 100.0 

Priors 0.6168 0.3832 



Valid8 WITHOUT TERMENLT 



From 

c1 1 2 Total 

3 1 4 

75.00 25.00 100.0 



3D 



1 58 8 
37.88 12.12 100.0 

2 39 10 49 
79.59 20.41 100.0 

Total 100 19 119 

Percent 34.03 15.97 100.0 

Priors 0.6168 0.3832 



65 



TABLE 4 3 

AW STEPWISE SELECTION: SUMMARY 
WITH TERMENLT 



Variable 
Entered 


Number 

In 


Partial 
R**2 


F 
Stat 


Prob 
F 


TERMENLT 

SCREEN 

ASVABAR 

ASVABSP 

ASVABSI 

ASVA3GS 


1 
2 
3 
4 
5 
6 


0.1881 
0.0064 
0.0061 
0.0037 
0.0035 
0.0039 


150.373 
4. 193 
3.982 
2.422 
2.270 
2.499 


0.0001 
0.0411 
0.0464 
0. 1202 
0. 1324 
0. 1 144 



TABLE 44 

AW STEPWISE SELECTION: SUMMARY 
WITHOUT TERMENLT 



Variable 
Entered 



Number 
In 



Partial 
R**2 



F 
Sra- 



Prob 
F 



SCREEN 

ASVABAR 

ASVABMK 

ENTRPAYG 



0.0158 
0.0072 
0.0060 
0.0043 



10.451 
4.723 
3.880 
2.791 



0.0013 
0.0301 
0.493 
0.0953 



66 



TABLE 45 
AW DISCRIMINANT ANALYSIS 

Deriv8 WITHOUT TERMENLT 



1 


1 


2 


Total 




100.0 


0.0 


100.0 


1 


439 
98.43 


7 
1.57 


446 
100.0 


2 


193 

94.15 


12 

5.85 


205 

100.0 


Total 
Percent 


552 
97.17 


19 

2.83 


571 

100.0 


Priors 


. 6 85 1 


0.3 149 





Valid8 WITHOUT TERMENLT 

C1 1 2 Total 

13 13 

100.0 00.0 100.0 

1 217 2 219 
99.09 .91 100.0 

2 100 3 103 
97.09 2.91 100.0 

Total 330 5 335 

Percent 98.51 1.49 100.0 

Priors 0.6851 0.3149 



67 



TABLE 46 
AX DISCRIMINANT ANALYSIS 

Deriv8 TERMENLT AS A VARIABLE 

From 

C1 1 2 Total 

7 5 12 

53.33 41.67 100.0 

1 130 28 158 
82.28 17.72 100.0 

2 34 62 96 
35.42 64.58 100.0 

Total 171 95 266 

Percent 64.29 35.71 100.0 

Priors .6220 .3780 



Valid8 TERMENLT AS A VARIABLE 



From 
C1 

• 


1 


00. 


2 

4 
100.00 


Total 

4 
100.0 


1 


47 
75.8 1 


15 
24. 19 


62 
100.0 


2 


12 
25.6 7 


33 
73.33 


45 

100.0 


Total 
Percent 


59 
53. 15 


52 

46.85 


111 

100.0 


Priors 


.6 220 


.378 





68 



TABLE 47 
AW DISCRIMINANT ANALYSIS 

Deriv8 TERMENLT AS A VARIABLE 

From 

C1 1 2 Total 

11 9 20 

55.0 45.0 130.0 

1 355 91 446 
79.60 20.40 130.0 

2 72 133 235 
35.12 64.88 133.3 

Total 430 233 671 

Percent 65.28 34.72 130.0 

Priors 0.6851 0.3 144 



Valid8 TERMENLT AS A VARIABLE 



From 
C1 

• 


1 
4 
30.77 


2 

9 

69.23 


Total 

13 
100.0 


1 


173 
79.00 


46 
21. 00 


219 
133.0 


2 


43 
41 .75 


60 
58. 25 


103 
133. 


Total 

Percent 


220 
65.67 


115 

34. 3 3 


3354 
100.0 


Priors 


0. 6 851 


0. 3149 





69 



LIST OF REFERENCES 



1. Vinberg, Robert and Joyner, John N. , Prediction of Job 
Performance: Review of Military Studies (NPRDC TR 82-37) , 
Navy Personnel Research and Development Center, 1982. 

2. Navy Enlisted Career Guide , 1980-1981. 

3 . Ibid. , p. 66 . 

4. Binkin, Martin and Back, Shirley J., Women and the Military , 
The Brookings Institution, 1977. 

5. Thomason, James S., First Term Survival and Reenlistment 
Changes for Navy Ratings and a Strategy for Their Use , 
1979. 

6. Office of the Assistant Secretary of Defense (Manpower, 
Reserve Affairs, and Logistics), ASVAB Working Group 
History of the Armed Forces Vocational Aptitude Battery 
(ASVAB) 1974-1980 , Washington, DC, 1980. 

7. Lurie, Philip M. , Continuous Estimates of Survival Through 
Eight Years of Service Using FY 1979 Cross-Sectional Data , 
1981. 

8. Lurie, Philip M. , Relating Enlistment Standards to Job 
Performance: A Pilot Study for Two Navy Ratings , 1981. 

9. Lockman, Robert F., First Term Success Predictions for 
Class A School and General Detail Recruits , 1973. 

10. Lurie, p. 56. 

11. Lockman, Robert F., The Effect of Delayed Entry, Recruit 
Quality and Training Guarantees of Two-Year Survival of 
Navy Enlisted Men , 1976. 

12. Thomason, p. 35. 

13. Marcus, Alan J., and Lockman, Robert F., Alternative 
Enlistment Standards , 1981. 

14. Lockman, Robert F. and Lurie, Philip M. , A New Look at 
Success Chances for Recruits Entering the Navy (SCREEN) , 
1980. 



70 



15. Sands, William A., Screening Male Applicants for Navy 
Enlistment ( NPRDC TR 77-34) , Navy Personnel Research and 
Development Center, San Diego, 1977. 

16 . Navy Enlisted Recruiting Manual . 

17. Kachigan, Sam K. , Multivariate Statistical Analysis A 
Conceptual Introduction , New York, 1982. 

18. Whitmire, Robert D. and Deitchman, Charles G. , An Enlisted 
Performance Prediction Model for the Aviation Structural 
Mechanics, M.S. Thesis, Naval Postgraduate School, 1983. 



71 



BIBLIOGRAPHY 



Bachman, Jerald G. , American High School Seniors View the 
Military, 1976-1982 , University of Michigan, 1982. 

Cory, Charles H., Predictive Validity of Surrogate Measures of 
Job Performance for Navy GENDETS , Navy Personnel Research and 
Development Center, San Diego, 1979. 

Fredlund, J. Eric and Little, Roger D. , Socioeconomic 
Characteristics of the All Volunteer Force: Evidence from the 
National Logitudinal Survey, 1979 , U.S. Naval Academy, 1982. 

Griffin, Patricia, A First Term Severity Index for U.S. Navy 
Ratings , M.S. Thesis, Naval Postgraduate School, 1981. 

Koehler, Ernest A., Life Cycle Navy Enlisted Billet Costs - 

FY 79 (NPRDC TR 79-13) , Navy Personnel Research and Development 

Center, San Diego, 1979. 

Koehler, Ernest A., Manpower Availability Projections for 
Selected Constrained Ratings - FY 1981-1987 (NPRDC TR 82-39) , 
Navy Personnel Research and Development Center, 1982. 

Lockman, Robert G. and O'Neill, David M. , Accession and 
Retention in the Navy, Summary Report 2 , Center for Naval 
Analysis, Alexandria, 1973. 

Lockman, Robert G. , Success Chances of Recruits Entering the 
Navy (CNS-1086) , Center for Naval Analysis, Arlington, 1977. 

Marcus, Alan J., Personnel Characteristics and Navy Aviation 
Squadron Performance: A Production Function Approach , Center 
for Naval Analysis, 1982. 

Military Manpower Task Force, A Report to the President on the 
Status and Prospects of the All Volunteer Force , U.S. Government 
Printing Office, 1982. 

Naval Manpower Research in the 1980 's Conference Proceedings 
(CNR 58) , Center for Naval Analysis, 1983. 

Navy Enlisted Initial Skill Rating Pipelines , CNET Notice 1514, 
Pensacola, 1981. 



72 



Office of the Assistant Secretary of Defense (Manpower, Reserve 
Affairs, and Logistics), Profile of American Youth: 1980 
Nationwide Administration of the Armed Services Vocational 
Aptitude Battery , U.S. Government Printing Office, 1982. 

Ruff, Richard R. , Shylo, Bruce J., Orth, Mollie N. , and 
Fraser, Jeannette L. , Military Training: Potential Roles for 
Vocational Education , Ohio State University, 1981. 

Rowe, Murry W. , "Manpower Models as Tools of Military Planning", 
Defense Management Journal , Fourth Quarter 1982. 

Thomas, George, The Feasibility of Modelling the Supply of 
Older Age Accessions , Naval Postgraduate School, 1982. 

Vineberg, Robert and Joyner, John N. , Performance of Men in 
Different Mental Categories , Human Resources Research Organiza- 
tion, 1978. 

Warner, John T. and Goldberg, Matthew S., The Influence of 
Non-Pecuniary Factors on Labor Supply: the Case of Navy 
Enlisted Personnel , Center for Naval Analysis, 1982. 

Warner, John T. and Thompson, Frederick, The Navy's Process 
for Planning Specialized Training (CNA 77-1230) , Center for 
Naval Analysis, 1977. 



73 



INITIAL DISTRIBUTION LIST 



No. Copies 



1 . Defense Technical Information Center 2 
Cameron Station 

Alexandria, Virginia 22314 

2. Library, Code 0142 2 
Naval Postgraduate School 

Monterey, California 93943 

3. Professor Thomas Swenson, Code 54Zw 1 
Naval Postgraduate School 

Monterey, California 93943 

4. Professor Richard Elster, Code 54Ea 2 
Naval Postgraduate School 

Monterey, California 93943 

5. Professor William McGarvey, Code 54 2 
Naval Postgraduate School 

Monterey, California 93943 

6. CDR Frank Keller, USN, Code 54 1 
Naval Postgraduate School 

Monterey, California 93943 

7. LCDR Clyde D. Sandel , USN 1 
Air Antisubmarine Squadron 32 

FPO New York 09501 

8. LT Mary F. Gleason, USN (Code 13) 1 
Commander Naval Telecommunications Command 

4401 Massachusetts Avenue N.W. 
Washington, DC 20390 



74 




>si 



Sandel 

lted + to Performance 
ln Av iation Antisub- 

^ne Warfare Operator 
and Avxation Antisub- 
marine Warfare Techni- 
cian ratings. 



'851 



The sis 
S15796 
c.l 



Sandel 

Enlisted standards as 
related to performance 
in Aviation Antisub- 
marine Warfare Operator 
and Aviation Antisub- 
marine Warfare Techni- 
cian ratings.